p-Terphenyl-Sensitized Photoreduction of CO2 with Cobalt and Iron Porphyrins. Interaction between CO and Reduced Metalloporphyrins

Article abstract of DOI:10.1021/jp991423u

Iron and cobalt porphyrins (FeP and CoP) are utilized as electron-transfer mediators to effect photochemical reduction of CO2 in homogeneous solutions. The species that activate and reduce CO2 are the Fe^0P and Co^0P formed by reduction of the starting materials. Reduction of the metalloporphyrins (MP) is achieved by photolysis in dimethylformamide or acetonitrile solutions containing triethylamine (TEA) as a reductive quencher. The photoreduction is efficient for the M^IIIP -> M^IIP stage and probably occurs by an intramolecular electron transfer from an axially bound TEA. However, TEA does not bind to the reduced metal complexes, and the quantum efficiency is much lower for the subsequent reduction steps. Considerably higher quantum yields are obtained by adding p-terphenyl (TP) as a sensitizer. TP is very effectively photoreduced by TEA to form the radical anion, TP(.-), which has a sufficiently negative reduction potential to reduce Co^IP and Fe^IP rapidly to their M^0P state. The rate constants for these reactions, determined by pulse radiolysis, are found to be nearly diffusion-controlled. The quantum yield for the reduction of M^IIP to M^IP and for reduction of CO2 to CO are increased by more than an order of magnitude in the presence of TP. Side reactions involve hydrogenation of the porphyrin ring and production of H2. The hydrogenated porphyrins also catalyze reduction of CO2. but the photochemical production of CO eventually stops. This limit on catalytic activity is due to destruction of the porphyrin macrocycle and accumulation of CO. CO can bind strongly to Fe^IIP and to Fe^IP but not to Fe^0P, as demonstrated by electrochemical measurements and by optical spectra of the species produced by sodium reduction in tetrahydrofuran in the presence and absence of CO. Although binding of CO to Fe^IIP and Fe^IP should not interfere with the formation of Fe^0P, the active catalyst, the potential for reduction of Fe^IP to Fe^0P becomes more negative. However, CO probably binds to the hydrogenated products thereby inhibiting the catalytic process.

Full text of DOI:10.1021/jp991423u

7742
J. Phys. Chem. A 1999, 103, 7742-7748
p-Terphenyl-Sensitized Photoreduction of CO2 with Cobalt and Iron Porphyrins. Interaction
between CO and Reduced Metalloporphyrins
T. Dhanasekaran, J. Grodkowski,† and P. Neta*
Physical and Chemical Properties DiVision, National Institute of Standards and Technology,
Gaithersburg, Maryland 20899-8381
P. Hambright
Department of Chemistry, Howard UniVersity, Washington, D.C. 20059
Etsuko Fujita*
Chemistry Department, BrookhaVen National Laboratory, Upton, New York 11973
ReceiVed: April 29, 1999; In Final Form: July 26, 1999
Iron and cobalt porphyrins (FeP and CoP) are utilized as electron-transfer mediators to effect photochemical
reduction of CO2 in homogeneous solutions. The species that activate and reduce CO2 are the Fe0P and Co0P
formed by reduction of the starting materials. Reduction of the metalloporphyrins (MP) is achieved by photolysis
in dimethylformamide or acetonitrile solutions containing triethylamine (TEA) as a reductive quencher. The
photoreduction is efficient for the MIIIP f MIIP stage and probably occurs by an intramolecular electron
transfer from an axially bound TEA. However, TEA does not bind to the reduced metal complexes, and the
quantum efficiency is much lower for the subsequent reduction steps. Considerably higher quantum yields
are obtained by adding p-terphenyl (TP) as a sensitizer. TP is very effectively photoreduced by TEA to form
the radical anion, TP•-, which has a sufficiently negative reduction potential to reduce CoIP and FeIP rapidly
to their M0P state. The rate constants for these reactions, determined by pulse radiolysis, are found to be
nearly diffusion-controlled. The quantum yield for the reduction of MIIP to MIP and for reduction of CO2 to
CO are increased by more than an order of magnitude in the presence of TP. Side reactions involve
hydrogenation of the porphyrin ring and production of H2. The hydrogenated porphyrins also catalyze reduction
of CO2, but the photochemical production of CO eventually stops. This limit on catalytic activity is due to
destruction of the porphyrin macrocycle and accumulation of CO. CO can bind strongly to FeIIP and to FeIP
but not to Fe0P, as demonstrated by electrochemical measurements and by optical spectra of the species
produced by sodium reduction in tetrahydrofuran in the presence and absence of CO. Although binding of
CO to FeIIP and FeIP should not interfere with the formation of Fe0P, the active catalyst, the potential for
reduction of FeIP to Fe0P becomes more negative. However, CO probably binds to the hydrogenated products
thereby inhibiting the catalytic process.
Introduction
factors that limit the catalytic activity of the present system were
also investigated.
Certain transition metal complexes act as electron transfer
mediators for photochemical1,2 or electrochemical3 reduction of
CO2. Numerous studies have been carried out on such systems
because of the interest in catalyzed reduction of CO2 as a means
of energy storage.4 Recent studies have shown that iron and
cobalt porphyrins are effective homogeneous catalysts for the
electrochemical and photochemical reduction of CO2 to CO and
formic acid.5-7 The mechanism was suggested to involve
binding of the CO2 to the reduced metalloporphyrin in its M0P
oxidation state. In these photochemical studies, the quantum
yields for the reduction of the metalloporphyrin and, conse-
quently, for the reduction of CO2, were relatively low. The
present study was aimed at exploring the use of p-terphenyl as
a photosensitizer for the metalloporphyrin-catalyzed reduction
of CO2 and at deriving the relevant kinetic parameters. The
Experimental Section8
Experiments were carried out with several cobalt and iron
porphyrins (MP). The ligands include 5,10,15,20-tetraphen-
ylporphyrins (TPP), tetrakis(3-methylphenyl)porphyrin (TTP),
tetrakis(3-fluorophenyl)porphyrin (T3FPP), tetrakis(3-trifluoro-
methylphenyl)porphyrin (T3CF3PP), tetrakis(pentafluorophen-
yl)porphyrin (TF5PP), and octaethylporphyrin (OEP). The
complexes were supplied by Mid-Century Chemicals (Posen,
IL) in the form of CoIIP and ClFeIIIP. The meta-substituted
derivatives were preferred to TPP in certain experiments because
they are more soluble in acetonitrile and in tetrahydrofuran
(THF). p-Terphenyl (TP, C6H5-C6H4-C6H5) from Aldrich was
recrystallized from alcohol. Triethylamine (TEA) was from
Aldrich, and acetonitrile (ACN) and N,N-dimethylformamide
(DMF) were analytical grade reagents from Mallinckrodt. TEA
and THF were purified as described previously.7
† On leave from the Institute of Nuclear Chemistry and Technology,
Warsaw, Poland.
10.1021/jp991423u CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/04/1999

Interaction between CO and Reduced Metalloporphyrins
J. Phys. Chem. A, Vol. 103, No. 38, 1999 7743
Irradiations were performed with fresh solutions that were
bubbled with Ar or He to remove oxygen or were saturated
with CO2. Photolysis was performed with a 300 W xenon lamp,
using a water filter to absorb the IR and a Pyrex filter to absorb
λ < 300 nm. Absorption spectra of the porphyrins were recorded
before and after irradiation, the CO evolved was determined
by gas chromatography (Carboxen-1000 column, thermal con-
ductivity detector), and the formate produced was analyzed by
a Dionex DX-500 ion chromatograph using an AS-11 column
and NaOH solutions as eluent. Formaldehyde was analyzed by
the chromotropic acid method,9 methanol, ethanol, and acetal-
dehyde were analyzed by GC/MS (Hewlett-Packard 5880A gas
chromatograph and 5970A mass spectrometer, using a Poraplot
Q capillary column at 150 °C) following distillation on a vacuum
line. Some of the metalloporphyrins were also reduced using a
sodium mirror in THF by standard vacuum line procedures using
home-built glassware.10 An excess of Na was generally used,
and the end point of each reduction was carefully monitored
by the loss of isosbestic points using a HP 8452A diode array
spectrophotometer.
On the other hand, photoreduction of MIIP and MIP by TEA
are less efficient bimolecular processes.
MIIP + Et3N
MIP- + Et3N
9
hν8 MIP- + Et3N•+
hν8 M0P2- + Et3N•+
(4)
(5)
9
Moreover, the Et2NC˙ HCH3 radical will reduce MIIP more slowly
than it reduces ClMIIIP, but it is not likely to reduce MIP-, since
the reduction potentials for MIP-/M0P2- (see below) are more
negative than that for the triethylamine radical.12
To increase the photochemical reduction yield, we explored
the use of p-terphenyl (TP) as a sensitizer.2 This compound has
been shown13 to form an excited state which is rapidly quenched
by TEA to produce the radical anion, TP•-.
TP
9
hν8 TP*
9
hν8 TP•- + Et3N•+
(6)
(7)
TP* + Et3N
To observe short-lived intermediates and determine rate
constants, pulse radiolysis was carried out with the apparatus
described previously,11 which utilizes 50 ns pulses of 2 MeV
electrons from a Febetron 705 pulser. Reaction rate constants
are reported with their estimated standard uncertainties. All
experiments were performed at room temperature, (20 ( 2) °C.
The reduction potential of the TP/TP•- couple is highly negative
(-2.45 V vs SCE in dimethylamine).13 Since the reduction
potentials for FeI/Fe0TPP (-1.64 V vs SCE)5 and CoI/Co0TPP
(-2.02 V vs SCE)7 are less negative, TP•- should reduce the
metalloporphyrins down to their M0P states.
Cyclic voltammograms were recorded with a BAS100B
electrochemical analyzer with scan rates ranging from 20 mV
s-1 to 1 V s-1. A conventional three-electrode system consisting
of a Pt or glassy carbon working electrode, a Pt counter
electrode, and a standard calomel reference electrode was used.
Ferrocene was added as an internal standard at the end of all
experiments. All potentials are given with reference to the
standard calomel electrode.
TP•- + MIIP f TP + MIP-
TP•- + MIP- f TP + M0P2-
(8)
(9)
Indeed, we find that the quantum yields for reduction of FeIIP
in DMF and CoIIP in acetonitrile, both containing 5% TEA,
are increased by more than an order of magnitude upon addition
of 3 mmol L-1 TP to the solution.
Continued photolysis in CO2-saturated solutions leads to
production of CO in a catalytic process. The total yield of CO
measured upon photolysis of 0.05 mmol L-1 FeIITPP in DMF/
TEA solutions in the presence of 3 mmol L-1 TP reached ∼4
mmol L-1, which is ∼6 times higher than that in the absence
of TP,6 and this yield was achieved ∼10 times faster than in
the absence of TP. A similar effect is found when comparing
the results for CoIITPP in ACN/TEA solutions in the absence
of TP (Figure 1a)7 with those obtained in the presence of TP.
Figure 1b shows the results for three concentrations of CoII-
T3FPP in the presence of 3 mmol L-1 TP. The initial rate of
CO production at these three concentrations is about 2 orders
of magnitude higher than in the absence of TP.
The considerable increase in the rate of photoproduction of
CO indicates that reactions 8 and 9 take place very efficiently.
To determine the rate constants for these reactions we used the
pulse radiolysis technique. Irradiation of a deoxygenated solution
of p-terphenyl (0.05 mol L-1) in DMF or DMSO leads to
formation of TP•-, which has several intense peaks between
400 and 500 nm and between 800 and 920 nm.13,14 By
monitoring the rate of decay at 480 or 840 nm as a function of
porphyrin concentration, we derived the rate constants for CoII-
TPP, k8 ) (7.4 ( 1.2) × 109 L mol-1 s-1 in DMSO and (1.0
( 0.2) × 1010 L mol-1 s-1 in DMF. The rate constants for
reaction 9 were determined in a similar manner except that the
solutions were first photolyzed to reduce the CoIITPP to
CoITPP- and then pulse irradiated to determine the decay rate
of TP•-. The values of k9 were found to be (6 ( 2) × 109 L
mol-1 s-1 in DMSO and (7 ( 2) × 109 L mol-1 s-1 in DMF.
Thus, both reactions 8 and 9 are essentially diffusion-controlled.
Results and Discussion
Kinetics and Mechanisms. The cobalt and iron porphyrins
were photochemically reduced in acetonitrile or DMF solutions
containing 5% TEA (0.36 mol L-1) as a reductive quencher, as
previously described.6,7 The quantum yield for reduction of FeIII-
TPP to FeIITPP was 0.2 under these conditions, whereas the
yields for reduction of MIIP to MIP are considerably lower, in
the range of 0.01 to 0.04. This difference was ascribed to the
fact that TEA binds as an axial ligand to MIIIP but it does not
bind to any significant extent to the MIIP or MIP complexes.6,7
In fact, the UV-vis spectra of FeIIP, FeIP, and Fe0P in TEA-
containing THF solutions are identical to the corresponding one
without TEA. Consequently, photochemical reduction of MIIIP
occurs by an efficient intramolecular electron transfer from the
bound TEA to the metal center,
Et3N(Cl)MIIIP
9
hν8 Et3N•+ + Cl- + MIIP
(1)
and is followed by reduction of another MIIIP molecule by the
reducing radical derived from TEA.
Et3N•+ + Et3N f Et3NH+ + Et2NC˙ HCH3
Et2NC˙ HCH3 + ClMIIIP f
Et2N+dCHCH3 + MIIP + Cl- (3)
(2)

7744 J. Phys. Chem. A, Vol. 103, No. 38, 1999
Dhanasekaran et al.
Figure 2. Photochemical production of HCOO- and CO in CO2-
saturated acetonitrile solutions containing 5% TEA and 3 × 10-3 mol
L-1 TP. (a) Yield of formate (O) without porphyrin; yield of formate
(b) and CO (2) with 1 × 10-4 mol L-1 CoTTP. (b) Yield of formate
(b) and CO (2) with 1 × 10-4 mol L-1 FeTTP. The solutions were
photolyzed in a 1 × 1 × 4 cm3 optical cell cooled in a water bath,
placed 10 cm away from the Xe lamp. The solution volume was 4 mL
and the headspace 2 mL.
Figure 1. Photochemical production of CO in CO2-saturated aceto-
nitrile solutions containing 5% TEA and Co porphyrin: (a) 1 × 10-5
mol L-1 CoIITPP with no TP; (b) with 3 × 10-3 mol L-1 TP and various
concentrations of CoIIT3FPP, (b) 9 × 10-5 mol L-1, (O) 2.4 × 10-5
mol L-1, (4) 7 × 10-6 mol L-1. The solutions were photolyzed in a
Pyrex bulb cooled by a water jacket, placed 10 cm away from an ILC
Technology LX-300 UV Xe lamp. The solution volume was 32 mL
and the headspace above the solution was 11 mL. The yield of CO in
mmol L-1 refers to the total amount of CO produced per unit volume
of solution.
adduct.
•CO2- + MIP- f (CO2MP)2-
(12)
This adduct is equivalent to that formed by reaction of M0P2-
with CO2,
Saturating these solutions with CO2 resulted in a lower
radiolytic yield of TP•- due to the rapid reaction of CO2 with
esol-, which competes with the reaction of TP with esol-, but
the rate of decay of the remaining TP•- appeared to be
unchanged, suggesting that reaction 10 is very slow (k10 e 106
L mol-1 s-1), as reported before.2b
M0P2- + CO2 f (CO2MP)2-
(13)
and will lead to formation of CO.
(CO2MP)2- + H+ f MIIP + CO + OH-
(14)
TP•- + CO2 f TP + •CO2
(10)
-
Photolysis of TP in ACN/TEA/CO2 in the absence of the
metalloporphyrin produces considerable yields of formate
(Figure 2a), in agreement with previous results.13 Under these
conditions, the TP is rapidly consumed and the production of
formate stops. In the presence of the metalloporphyrin, however,
the TP is not consumed because reactions 8 and 9 predominate.
The yield of formate in the presence of metalloporphyrins is
lower than that without metalloporphyrins at short photolysis
times, but reaches the same or higher values at longer times.
Parts a and b of Figure 2 show the results obtained with CoTTP
and FeTTP, respectively. Similar results were obtained with
CoT3FPP. The initial decrease in the yield of formate upon
addition of the porphyrin to the TP solution is accompanied by
a large increase in the yield of CO, which is not produced at all
in the absence of the metalloporphyrin. Comparison of the initial
slopes of the curves in parts a and b of Figure 2 also shows
that the initial rates of photochemical production of CO and
formate with FeTPP and with CoTPP are similar.
Nevertheless, this reaction may be expected to take place under
certain conditions, since the reduction potential of CO2 (-2.1
V vs SCE in water)15 is less negative than that of TP. Despite
its low rate constant, reaction 10 may have a contribution in
the typical photochemical experiment, since the concentration
of CO2 is at least 3 orders of magnitude higher than the
concentration of the metalloporphyrin. It is likely, however, that
most of the CO2 formed by reaction 10 will react with the
metalloporphyrin. Reaction with MIIP will take place with a
rate constant of the order of 108 L mol-1 s-1 16 and will lead to
reduction.
•
-
•CO2- + MIIP f CO2 + MIP-
(11)
Reaction with MIP- may occur more slowly7 and form an

Interaction between CO and Reduced Metalloporphyrins
J. Phys. Chem. A, Vol. 103, No. 38, 1999 7745
a
TABLE 1: Photochemical Production of CO and H2
Factors Limiting the Yields of Products. Catalytic pho-
toreduction of CO2 to CO and formate by the metalloporphyrins,
as seen in Figures 1 and 2 and in many other experiments, slows
down and eventually stops after extended photolysis. It should
be noted that the original color of the solution changes in the
early stages of the photolysis, turning to green (chlorin) after
only 10-20% of the CO and formate are produced, and later
turning yellowish. Formation of CO and formate continues even
after the solution becomes colorless, until eventually the
concentrations of products level off. The limit on production
of CO and formate may be the result of complete degradation
of the porphyrin (see below).
initial rate (mmol L-1 h-1
)
yieldb (mmol L-1
)
compound
CO
H2
CO
H2
CoTTP
FeTTP
0.45
0.84
0.2
0.1
3.1
2.1
1.6
3.4c
a Measured in CO2-saturated acetonitrile solutions containing 5%
TEA, 3 mmol L-1 TP, and ∼0.05 mmol L-1 metalloporphyrin. The
solution volume was 35 mL and the headspace was 8.4 mL. The
estimated standard uncertainties are (10% for the values of CO and
(20% for H2. b Measured after ∼20 h photolysis, when the concentra-
tions of products generally leveled off. c Production of H2 continued
while production of CO stopped.
The total amount of CO formed was dependent on the
porphyrin concentration (Figure 1b) but not on the TP concen-
tration (varied from 1 to 8 mmol L-1). This indicates that the
porphyrin (e0.1 mmol L-1) is depleted, while the TP (>1 mmol
L-1) and certainly TEA (0.36 mol L-1) and CO2 (∼0.2 mol
L-1) are not depleted. We then compared several complexes:
FeTPP, FeTF5PP, FeOEP, and FeOECn (octaethylcorrphycene).
By using 1 × 10-4 mol L-1 of each metalloporphyrin and 3
mmol L-1 TP in CO2-saturated DMF/TEA solutions, we found
that the initial rate of photochemical production of CO was
similar for all four complexes and the total yield of CO leveled
off at ∼4 mmol L-1 for the TPP and OEP and ∼3 mmol L-1
for the other two complexes. This indicates that the limitation
in the CO production is not strongly related to the structure or
the reduction potential of the porphyrin. Comparison of parts a
and b of Figures 2 shows that the total yield of CO obtained
with CoTTP is ∼1.5 times higher than that obtained with FeTTP,
although the total yields of formate were comparable with the
two porphyrins. As noted above, however, all the Fe and Co
porphyrins studied undergo complete bleaching during the
photochemical reduction of CO2 and yet the catalytic process
still continues to produce CO for an extended period. It was
further noticed that the colorless photolyzed solutions, when
exposed to O2 for several days, became brown-red. The spectra
of these oxidized products, however, were different than the
spectra of the original porphyrins. They showed a strong peak
at 418 nm and a long tail up to 700 nm but no additional peaks.
TLC analysis of the products detected no metalloporphyrin.
Electrospray mass spectrometric analysis detected little of the
original metalloporphyrin (FeTPP, m/z ) 668) but mostly
fragments (m/z < 400).
These results indicate that the porphyrin macrocycle is
decomposed during the photochemical reduction of CO2. The
observed bleaching is due to a side reaction and arises from
protonation of the macrocycle of reduced intermediates. Reduc-
tion of MIP- produces M0P2-, in which part of the electron
density resides on the macrocycle π-system. Reaction of M0P2-
with CO2 leads to production of CO. Reaction of M0P2- with
protons, formed via reaction 2, may lead to protonation of the
metal center and eventual formation of H2.17 In fact, H2 is
detected as an additional product along with CO in all the current
experiments. The yield of H2 was often lower than the yield of
CO, but the ratio varied greatly with porphyrin (Table 1) and
with conditions (detailed results will be presented in an
upcoming publication). In addition to protonation at the metal,
partial protonation at the meso or pyrrolic carbons apparently
takes place to produce metal complexes with reduced macro-
cycles. Chlorins are observed among the early products, and it
was shown before6 that iron tetraphenylchlorin also catalyzes
the photochemical reduction of CO2 to CO. Further saturation
of pyrrole double bonds will produce bacteriochlorins and
isobacteriochlorins, which have significant visible absorptions,
but saturation of the meso positions eventually produces
porphyrinogens,18 which are colorless. It is likely that any of
these complexes can be reduced by TP•- to form a species that
can catalyze reduction of CO2. It is known that the colorless
porphyrinogens can be oxidized by O2 to the corresponding
porphyrins but that products saturated at the pyrrole double
bonds are not readily oxidized by O2. The fact that the oxidized
products in our experiments are only fragments of the original
porphyrins may be due to fragmentation of the macrocycle
during the reduction process19 or during the slow air-oxidation
process.
If the hydrogenated metalloporphyrins continue to catalyze
the reduction of CO2, then the limit on the CO yield may be
due also to accumulation of products. Two products accumulate
in these solutions which may have an effect on the catalytic
reaction, namely CO and acetaldehyde. Acetaldehyde is pro-
duced by oxidation of TEA and hydrolysis of the imine product20
(formed for example by reaction 3). Product analysis after
extensive photolysis showed the presence of acetaldehyde as
well as ethanol. This indicates that part of the acetaldehyde
produced in this system is reduced to ethanol. In principle,
acetaldehyde may be reduced by TP•- or by M0P2-. We carried
out pulse radiolysis experiments, similar to those described
above, and found that the lifetime of TP•- in DMF is not
significantly reduced by the addition of 0.05 mol L-1 acetal-
dehyde. We conclude, therefore, that the reaction of acetalde-
hyde with TP•- is very slow (k e 106 L mol-1 s-1) and can be
neglected in this system.21 In support of this conclusion, we
also find that 0.05 mol L-1 acetaldehyde does not affect the
quantum yield for photoreduction of FeIIP to FeIP- in DMF/
TEA/TP solutions. However, addition of 0.05 mol L-1 acetal-
dehyde decreases the rate of production of CO by about a factor
of 2. Since acetaldehyde is continuously produced in the
photolysis, it will accumulate to a steady-state concentration at
which its production rate will equal the rate of subsequent
reduction. Thus, accumulation of acetaldehyde constrains CO
production but is not the major limiting factor, and the results
described below indicate that accumulation of CO is the critical
limiting factor.
To examine the effect of the CO accumulated in solution,
we diminished the CO concentration by increasing the headspace
above the solutions. We photolyzed identical volumes of a
solution of FeTPP in DMF/TEA/TP/CO2 in cells with different
total volumes. We find that by increasing the headspace above
the solution the catalytic reduction of CO2 continues for longer
periods and yields larger amounts of CO (Figure 3). This
indicates that if CO is allowed to escape from the solution the
catalytic process will continue. In support of this conclusion,
we also find that when the above photolyzed solutions are
purged with CO2 again, to remove the accumulated CO, the
photochemical production of CO is resumed. Inversely, if a
certain amount of CO is introduced into the solution before

7746 J. Phys. Chem. A, Vol. 103, No. 38, 1999
Dhanasekaran et al.
TABLE 2: Reduction Potentials of Iron Porphyrins in
n-Butyronitrile
porphyrin
potentialsa vs SCE
Ip(CO2)/Ip(Ar)b
ClFeIIITPP
-0.26, -1.05, -1.66
-0.23, -1.02, -1.61
-0.19, -1.00, -1.55
2.0
1.3
1.2
ClFeIIIT3CF3PP
ClFeIIIT3FPP
a Potentials for FeIII/FeII, FeII/FeI, and FeI/Fe0, with estimated standard
uncertainties of (0.01 V. b Estimated standard uncertainties (15%.
TABLE 3: Spectroscopic Properties of Iron Porphyrins in
THFa
porphyrin
ClFeIIITPP
λ )
max, nm (10-3 ꢀ, L mol-1 cm-1
278 (21.2), 370 (55.4), 416 (112), 506 (13.6),
574 (4.0), 660 (3.0), 688 (3.6)
252 (20.4), 294 sh (22.8), 308 (23.5), 426 (241),
540 (11.0), 554 sh (10.5), 598 sh (4.2), 684 sh (1.1)
258 (26.9), 314 (18.5), 414, 534 (12.2),
570 sh (4.3), 610 (2.8)
282 (27.4), 330 (32.8), 394 (67.9), 424 (76.1),
512 (13.8), 572 (7.9), 676 (4.4), 712 sh (3.4)
254 (27.2), 298 sh (25.0), 358 (66.8), 450 (68.8),
514 (25.2), 710 (5.4), 778 (7.8)
FeIITPP
(CO)FeIITPP
FeITPP-
Figure 3. Photochemical production of CO in CO2-saturated solutions
containing 8.6 × 10-5 mol L-1 FeTPP, 3 × 10-3 mol L-1 TP, and 5%
TEA in DMF. The solution volume was 32 mL and the headspace above
the solutions was varied: 11 mL (b); 28 mL (O); 52 mL (4).
Fe0TPP2-
ClFeIIIT3CF3PP
FeIIT3CF3PP
278 (22.2), 364 (52.8), 414 (109), 505 (12.8),
572 (4.5), 650 (3.2), 678 (3.1)
262 (18.3), 308 (23.6), 366 sh (19.4), 428 (230),
542 (10.2), 600 sh (3.5)
256 sh (30.5), 320 (25.8), 416 (244), 532 (14.7)
286 (20.0), 330 (29.5), 394 (44.1), 428 (74.3),
514 (11.5), 578 sh (6.8), 680 (2.7), 712 sh (2.4)
photolysis, the amount of CO produced by the photolysis is
correspondingly diminished.
(CO)FeIIT3CF3PP
FeIT3CF3PP-
We then examined whether the concentration of CO levels
off because this product is reduced to formaldehyde or methanol
in competition with reduction of CO2. We analyzed the solutions
after extensive photolysis but detected no formaldehyde or
methanol. We conclude, therefore, that the accumulated CO may
interfere with the reduction of CO2 by binding to the metal ion
and preventing the binding of CO2. It has been suggested before
that CO binds to Fe(I) porphyrins22,23 and that the resulting
complex,22 upon further reduction, is more likely to undergo
protonation at the meso position to form the phlorin anion.
Consecutive reductions of such complexes in our system may
explain the extensive formation of ligand-reduced products.
Binding of CO to these reduced complexes may compete with
the binding of CO2 and thus prevent its reduction.
(CO)FeIT3CF3PP- 254 sh (22.8), 324 (20.6), 396 sh (25.8),
420 sh (46.0), 435 (78.1), 450 sh (46.0), 662 (7.3)
Fe0T3CF3PP2-
256 sh (19.9), 362 (42.6), 452 (51.2), 516 (17.2),
720 sh (3.5), 792 (4.4)
a Molar absorption coefficients ꢀ were calculated by assuming 100%
conversion from the ClFeIIIP species. Their estimated standard uncer-
tainties are (10%.
Interaction of CO with the Reduced Metalloporphyrins.
To further investigate the effect of CO in these systems, we
carried out reduction of iron porphyrins by electrochemical
methods and by sodium. Cyclic voltammetry and differential
pulse polarography of ClFeIIITPP in Ar-saturated butyronitrile
solutions show peaks for three one-electron reduction steps, FeIII/
FeIIP, FeII/FeIP, and FeI/Fe0P, at -0.26, -1.05, and -1.66 V
vs SCE, respectively. Identical reduction peaks are observed in
CO2-saturated solutions; catalytic current is observed at the FeI/
Fe0P reduction step. However, when the solutions were saturated
with CO, the FeIII/FeIIP peak became more positive, -0.19 V,
the FeII/FeIP became more negative, -1.29 V, but the third peak
remained unchanged. These shifts are in agreement with
previous results23,24 and indicate that CO binds to FeIIP and to
FeIP, but not to Fe0P. Similar changes take place when THF is
used as solvent. Under Ar and CO2, the potentials for FeIII/
FeIIP, FeII/FeIP, and FeI/Fe0P are -0.29 (irr), -1.12, and -1.64
V, respectively. Under CO, they are at -0.3 (irr), -1.35, and
-1.64 V vs SCE, respectively. It should be emphasized that
CO binding to FeIP is demonstrated here for unsubstituted TPP
in a nonpolar noncoordinating solvent. Addition of TEA to the
THF solution does not change the potentials, which indicates
almost no interaction between TEA and FeTPP in any oxidation
state with and without CO or CO2. Similar results were obtained
with ClFeIIIT3FPP and ClFeIIIT3CF3PP (Table 2).
Figure 4. Spectral changes observed upon sodium reduction in THF:
(a) reduction of ClFeIIIT3CF3PP; (b) reduction of FeIIT3CF3PP; (c)
reduction of FeIT3CF3PP-.
solutions of ClFeIIITPP and ClFeIIIT3CF3PP under vacuum. The
results are summarized in Table 3. Figure 4a shows the stepwise
reduction of ClFeIIIT3CF3PP (Soret peak at 414 nm) to FeII-
T3CF3PP (Soret peak at 428 nm). Figure 4b shows the stepwise
reduction of the reddish solution of FeIIT3CF3PP to the greenish-
brown solution of FeIT3CF3PP- (split Soret at 394 and 428 nm).
To examine this binding through optical absorption spectra,
we carried out stepwise reduction by a sodium mirror in THF

Interaction between CO and Reduced Metalloporphyrins
J. Phys. Chem. A, Vol. 103, No. 38, 1999 7747
Figure 6. Spectral changes observed with a THF solution of Fe0TPP2-
following exposure to 760 Torr of CO2. The spectra were recorded at
2, 4, 6, 9, 13, 36, and 150 min after the addition of CO2.
Similar experiments with CoIITPP and CoIIT3CF3PP indicated
that CO does not bind to CoIIP, CoIP-, or Co0P2-. This is in
contrast to the more saturated tetraazamacrocyclic complexes,
where the CoI complexes are known to bind CO.26 The above
results indicate that CO does not interfere with the ability of
Co0P2- and Fe0P2- to reduce CO2, but after the macrocycle is
hydrogenated CO may exert a considerable effect on the
reactivities of those complexes.
Figure 5. Absorption spectra of FeIIP and FeIP in THF solutions in
the absence and presence of CO. (a) FeIIT3CF3PP (dotted line) and
(CO)FeIIT3CF3PP (solid line). (b) FeIT3CF3PP- at room temperature
(dotted line), (CO)FeIT3CF3PP- at room temperature (dashed/dotted
line), and (CO)FeIT3CF3PP- at low temperature (solid line). The
FeIT3CF3PP- solution contains trace amounts of Fe0T3CF3PP2-
(shoulder at 450 nm) and of (CO)FeIIT3CF3PP (shoulder at 416 nm).
Addition of CO2 to a THF solution containing FeIP- does
not cause any spectral change. However, addition of CO2 to a
solution of Fe0P2- immediately produces a (CO)FeIIP-like
spectrum, with shoulders at 370 and 450 nm, which may be
due to the existence of a small amount of Fe0P2- or (CO2)Fe0P2-.
Within 10 min, the intensities of the shoulders diminish and
the intensity at 414 nm reaches its maximum, indicating the
formation of (CO)FeIIP, as shown in Figure 6. In the next 2 h,
the intensity at 414 nm remains the same but a peak at 426 nm
appears and increases, indicating the formation of FeIIP. The
final solution contains FeIIP and (CO)FeIIP as an equilibrium
mixture. Attempts to produce and characterize the (CO2)FeTPP2-
complex by laser flash photolysis failed because of a low
quantum yield for the formation of Fe0TPP2- in the FeITPP-/
CO2/TEA/THF system. Addition of TP to the system increased
the quantum yield of Fe0TPP2-, but it caused other complica-
tions due to the formation of iron chlorin. Experiments using
the more stable phthalocyanines are underway.
Figure 4c shows the stepwise reduction of FeIT3CF3PP- to
brownish Fe0T3CF3PP2- (split Soret peaks at 362 and 452 nm).
These changes are similar to those reported for reduction of
ClFeIIITPP.7,25 Reduction in the presence of TEA produced an
identical spectrum at each reduction step, indicating that the
interaction of TEA with all oxidation states of FeTPP is not
strong.
Addition of CO to solutions of FeIIT3CF3PP and FeIT3CF3PP-
resulted in spectral changes (Figure 5 and Table 3) that indicate
formation of (CO)FeIIT3CF3PP (Soret peak at 416 nm) and
(CO)FeIT3CF3PP (Soret peak at 435 nm). Addition of CO to a
solution of Fe0T3CF3PP2-, however, did not lead to any change
in the absorption spectrum even after 2 h, suggesting that this
compound does not bind or react with CO. In certain experi-
ments, when the FeIIP was not thoroughly reduced to FeIP, a
second peak in the Soret region at 416 nm, with varying
intensities, was also observed and was clearly due to (CO)FeII-
T3CF3PP formed along with (CO)FeIT3CF3PP-. The solution
of (CO)FeIT3CF3PP- can also be obtained from the solution
of (CO)FeIIT3CF3PP by Na mirror reduction under CO. The
solution of (CO)FeIT3CF3PP- is thermochromic, brown at room
temperature and olive green at low temperature, with the main
peak at 436 nm. The spectrum changed reversibly upon raising
and lowering the temperature. At room-temperature both
FeIT3CF3PP- and (CO)FeIT3CF3PP- exist as an equilibrium
mixture ((CO)FeIT3CF3PP-/FeIT3CF3PP ≈ 1 under 1 atm of
CO). At low temperature, the solution contains only (CO)FeI-
T3CF3PP-. The binding of CO was fully reversible and the
spectrum reverted to that of FeIP- upon removal of the CO.
Further sodium reduction of FeIP- under vacuum and of
(CO)FeIP- under CO yielded the same product, Fe0P2-. This
leads to the conclusion that CO does not bind to Fe0P2-. Under
these thoroughly aprotic conditions, the spectrum of the Fe0P2-
was identical with or without CO and there was no evidence of
phlorin anion formation (observed before22 under less thoroughly
aprotic conditions). The above spectra of the iron porphyrins,
with or without CO, were unaffected by the addition of TEA.
Summary and Conclusions
Photochemical reduction of CO2 to CO is catalyzed by iron
and cobalt porphyrins in homogeneous solutions. The metal-
loporphyrins are reduced to their M0P2- state by photolysis in
organic solvents containing TEA, and the M0P2- species reduce
the CO2 to CO. The quantum yields for the photoreduction of
the metalloporphyrin and of the CO2 are greatly increased in
the presence of p-terphenyl. This compound is photoreduced
by TEA very effectively to produce a radical anion, which is
found to reduce the metalloporphyrins with diffusion-controlled
rate constants down to their M0P2- state. Side reactions in these
systems lead to hydrogenation of the porphyrin macrocycle, but
these reduced complexes also act as catalysts for reduction of
CO2 until they are decomposed. Acetaldehyde, formed in these
solutions by photooxidation of TEA, is found to be reduced to
ethanol, and this reduction is in competition with the reduction
of CO2. Accumulation of CO in the solution imposes a limit
on further reduction of CO2. The inhibition effect of CO
probably results from CO binding to the metal center thereby
competing with the binding of CO2. Since CO binds strongly

7748 J. Phys. Chem. A, Vol. 103, No. 38, 1999
Dhanasekaran et al.
to FeIIP and FeIP- but not to Fe0P2-, CoIIP, or its reduced states,
any Fe0P2- or Co0P2- formed in the solution can react rapidly
with CO2 to form CO. In competition with this process, Fe0P2-
and Co0P2- can react with the protons formed by the photolytic
reaction. This leads to partial formation of H2 and partial
hydrogenation of the porphyrin ligand. Both of these products
are observed. Further hydrogenation of the ligand makes the
reduction potentials of the complexes more negative and changes
their behavior. It may be speculated that some of these
hydrogenated complexes may react with CO2 in their FeI and
CoI oxidation states, instead of the Fe0 and Co0 states. Such
behavior would be comparable to that observed with the cobalt
complexes derived from cyclam and similar ligands.2 CO
binding to the hydrogenated FeI and CoI complexes may inhibit
further CO2 reduction. Future studies are aimed at examining
the properties of the hydrogenated complexes.
1988. Behr, A. Carbon Dioxide ActiVation by Metal Complexes; VCH:
Weinheim, 1988. Electrochemical and Electrocatalytic Reactions of Carbon
Dioxide; Sullivan, B. P., Ed.; Elsevier: Amsterdam, 1993. Sutin, N.; Creutz,
C.; Fujita, E. Comments Inorg. Chem. 1997, 19, 67.
(5) Hammouche, M.; Lexa, D.; Save´ant, J.-M.; Momenteau, M. J.
Electroanal. Chem. Interfacial Electrochem. 1988, 249, 347. Hammouche,
M.; Lexa, D.; Momenteau, M.; Save´ant, J.-M. J. Am. Chem. Soc. 1991,
113, 8455. Bhugun, I.; Lexa, D.; Save´ant, J.-M. J. Am. Chem. Soc. 1994,
116, 5015. Bhugun, I.; Lexa, D.; Save´ant, J.-M. J. Am. Chem. Soc. 1996,
118, 1769.
(6) Grodkowski, J.; Behar, D.; Neta, P.; Hambright, P. J. Phys. Chem.
A 1997, 101, 248.
(7) Behar, D.; Dhanasekaran, T.; Neta, P.; Hosten, C. M.; Ejeh, D.;
Hambright, P.; Fujita, E. J. Phys. Chem. 1998, 102, 2870.
(8) The mention of commercial equipment or material does not imply
recognition or endorsement by the National Institute of Standards and
Technology, nor does it imply that the material or equipment identified are
necessarily the best available for the purpose.
(9) Mitchell, J., Jr.; Kolthoff, I. M.; Proskauer, E. S.; Weissberger, A.
Organic Analysis; Interscience: New York, 1961; Vol. 1, p 286.
(10) Shriver, D. F.; Drezdzen, M. A. The Manipulation of Air-SensitiVe
Compounds, 2nd ed.; Wiley: New York, 1986. Kobayashi, H.; Hara, T.;
Kaizu, Y. Bull. Chem. Soc. Jpn. 1972, 45, 2148.
(11) Neta, P.; Huie, R. E. J. Phys. Chem. 1985, 89, 1783.
(12) Wayner, D. D. M.; McPhee, D. J.; Griller, D. J. Am. Chem. Soc.
1988, 110, 132. Armstrong, D. A.; Rauk, A.; Yu, D. J. Am. Chem. Soc.
1993, 115, 666.
(13) Matsuoka, S.; Kohzuki, T.; Pac, C.; Ishida, A.; Takamuku, S.;
Kusaba, M.; Nakashima, N.; Yanagida, S. J. Phys. Chem. 1992, 96, 4437.
(14) Shida, T. Electronic Absorption Spectra of Radical Ions; Elsevier:
New York, 1988; p 141.
(15) Schwarz, H. A.; Dodson, R. W. J. Phys. Chem. 1989, 93, 409.
(16) Baral, S.; Neta, P. J. Phys. Chem. 1983, 87, 1502.
(17) Bhugun, I.; Lexa, D.; Save´ant, J.-M. J. Am. Chem. Soc. 1996, 118,
3982.
(18) Fuhrhop, J.-H. In Porphyrins and Metalloporphyrins; Smith, K.
M., Ed.; Elsevier: Amsterdam, 1975; Chapter 14, p 614.
(19) Chatt, J.; Elson, C. M.; Leigh, G. J. J. Am. Chem. Soc. 1973, 95,
2408.
(20) DeLaive, P. J.; Foreman, T. K.; Giannotti, C.; Whitten, D. G. J.
Am. Chem. Soc. 1980, 102, 5627.
(21) The finding that both acetaldehyde and CO2 are reduced very slowly
by TP•-, while the porphyrins are reduced very rapidly, is probably related
to the fact that these small molecules undergo a change in geometry upon
reduction.
(22) Balducci, G.; Chottard, G.; Gueutin, C.; Lexa, D.; Save´ant, J.-M.
Inorg. Chem. 1994, 33, 1972.
(23) Swistak, C.; Kadish, K. M. Inorg. Chem. 1987, 26, 405.
(24) Croisy, A.; Lexa, D.; Momenteau, M.; Saveant, J.-M. Organome-
tallics 1985, 4, 1574.
Acknowledgment. This research was supported by the
Division of Chemical Sciences, Office of Basic Energy Sciences,
U.S. Department of Energy, under Contracts DE-AI02-
95ER14565 (NIST) and DE-AC02-76CH00016 (BNL). P.H.
thanks the Howard University CSTEA project (NASA contract
NCC S-184) for financial support. We thank Dr. Bruce
Brunschwig (BNL) for helpful discussions and for his participa-
tion in the preliminary laser photolysis experiments.
References and Notes
(1) Hawecker, J.; Lehn, J.-M.; Ziessel, R. J. Chem. Soc., Chem.
Commun. 1985, 56. Grant, J. L.; Goswami, K.; Spreer, L. O.; Otvos, J. W.;
Calvin, M. J. Chem. Soc., Dalton Trans. 1987, 2105. Craig, C. A.; Spreer,
L. O.; Otvos, J. W.; Calvin, M. J. Phys. Chem. 1990, 94, 7957. Kelly, C.
A.; Mulazzani, Q. G.; Venturi, M.; Blinn, E. L.; Rodgers, M. A. J. J. Am.
Chem. Soc. 1995, 117, 4911.
(2) (a) Fujita, E.; Brunschwig, B. S.; Ogata, T.; Yanagida, S. Coord.
Chem. ReV. 1994, 132, 195. (b) Ogata, T.; Yanagida, S.; Brunschwig, B.
S.; Fujita, E. J. Am. Chem. Soc. 1995, 117, 6708. (c) Matsuoka, S.;
Yamamoto, K.; Ogata, T.; Kusaba, M.; Nakashima, N.; Fujita, E.; Yanagida,
S. J. Am. Chem. Soc. 1993, 115, 601. (d) Ogata, T.; Yamamoto, Y.; Wada,
Y.; Murakoshi, K.; Kusaba, M.; Nakashima, N.; Ishida, A.; Takamuku, S.;
Yanagida, S. J. Phys. Chem. 1995, 99, 11916.
(3) Fisher, B.; Eisenberg, R. J. Am. Chem. Soc. 1980, 102, 7361.
Sullivan, B. P.; Bolinger, C. M.; Conrad, D.; Vining, W. J.; Meyer, T. J. J.
Chem. Soc., Chem. Commun. 1985, 1414. Beley, M.; Collin, J.-P.; Ruppert,
R.; Sauvage, J.-P. J. Am. Chem. Soc. 1986, 108, 7461. Fujita, E.; Haff, J.;
Sanzenbacher, R.; Elias, H. Inorg. Chem. 1994, 33, 4627.
(25) Anxolabehere, E.; Chottard, G.; Lexa, D. New J. Chem. 1994, 18,
889.
(4) Catalytic ActiVation of Carbon Dioxide; Ayers, W. M., Ed.; ACS
Symposium Series 363; American Chemical Society: Washington, DC,
(26) Fujita, E.; Creutz, C.; Sutin, N.; Szalda, D. J. J. Am. Chem. Soc.
1991, 113, 343.